Driveway Spacing and Traffic Operations

Driveway Spacing and Traffic Operations ABSTRACT JEROME S. GLUCK, GREG HAAS, JAMAL MAHMOOD Urbitran Associates 71 West 23rd Street, 11th Floor New York, NY 10010 urbitran@ix.netcom.com HERBERT S. LEVINSON Transportation Consultant 40 Hemlock Street New Haven, CT 06515 This paper reviews research studies relating traffic operations to access spacing, presents results of specially conducted operations analyses at 22 sites in Connecticut, Illinois, New Jersey, and New York, and sets forth emergent access spacing guidelines. The literature review and operational analyses were performed as part of NCHRP Project 3-52 Impacts of Access Management Techniques. Each site represented an unsignalized driveway for a major traffic generator along a suburban arterial roadway without deceleration lanes. Information was gathered on the number and percentage of through vehicles impacted by right turns. The percentage of through vehicles impacted approximated 0.18 times the rightturn volumes. The impact lengths of through vehicles were determined, and the influence distances were computed. The results were then used to quantify the likely effects of multiple driveways and to establish guidelines for deceleration lanes and access separation distances. Access separation distances for various operating speeds and right-turn volumes were based on the likelihood of minimizing spillback across an upstream driveway over a series of driveways along a 1 4 -mile section of road. For example, to hold the spillback rate to 20% for a 40-mph posted speed, a 285-foot spacing would be needed. When the acceptable spillback rate is reduced to 5%, a 400-foot separation is required. DISCLAIMER The opinions and conclusions expressed or implied in this report are those of the research agency that performed the research, and, while they have been accepted as appropriate by the technical committee, they are not necessarily those of the Transportation Research Board, the National Research Council, the American Association of State Highway and Transportation Officials, or the Federal Highway Administration, U.S. Department of Transportation. INTRODUCTION The spacing, location and design of access points influence traffic operations and safety. Too many closely spaced street and driveway intersections increase delays, while too few may inhibit access and over-concentrate traffic. C-3 / 1

C-3 / 2 TRB Circular E-C019: Urban Street Symposium This paper reviews research studies relating traffic operation and access spacing, presents the results of specially conducted operational analysis, and sets forth emergent guidelines for unsignalized access spacing. The literature review and operational analyses were performed as part of NCHRP Project 3-52 Impacts of Access Management Techniques (1). OVERVIEW OF OPERATIONS EXPERIENCE Several operational studies have attempted to quantify the travel time impacts associated with access spacing through field observations or traffic simulations. There were five studies conducted between 1962 and 1978, and four more recent studies. The studies found that increasing the number of driveways (i.e., reducing driveway spacing) along a section of highway increased delays and reduced roadway capacities. The methods and results varied from study-to-study, and there were no before and after studies. The field studies by Reilly et al. (7 ) and the simulations by McShane (11) gave generally consistent results. For driveway volumes of 600 per hour per mile, the Reilly studies (for uninterrupted flow) showed a speed loss of 1.0 to 1.7 mph per driveway, while the McShane simulations (for signalized arterials) suggested a 1.0 to 2.0 mph loss per driveway. OPERATIONS ANALYSIS Special studies were conducted to identify how right turns entering a driveway affect other drivers following in the same travel lane. The number of incidents was observed when brake lights were activated or evasive maneuvers were taken by a through vehicle following a right-turning vehicle. These observations served as surrogate measures for the number of impacts. The field investigations and analyses were conducted for 22 sites in Connecticut, Illinois, New Jersey, and New York. Each site represented a major traffic generator along a suburban arterial roadway. The arterials had no deceleration lane, and the driveways were not signalized. Information was gathered on the number and percentage of through vehicles impacted by right turns. The impact lengths of through vehicles impacted were determined, and, in turn, influence areas were computed. The results were used to quantify the effects of multiple driveways and to develop inputs for establishing unsignalized access spacing guidelines. 1. The number and percentage of through vehicles in the right lane that were impacted by right-turn-in at a single driveway. 2. The percentage of through vehicles in the right lane that were impacted by rightturn-in over a series of driveways. 3. The distances back from a single driveway entrance that cars began to be impacted the impact length and the spatial distributions of impacted vehicles. 4. The influence areas or influence distances before (upstream of) a driveway entrance. This involved adding perception-reaction distance and car length to the impact length.

Gluck, Haas, Mahmood, and Levinson C-3 / 3 5. The proportions of through vehicles in the right lane whose influence lengths extended to or beyond at least one upstream driveway over a section of road (spillback rate). 6. The variations of spillback rate by roadway operating speed. Through Vehicles Impacted by Right Turns The number and percentage of right (curb) lane through vehicles impacted by vehicles turning right into an unsignalized driveway were obtained from field measurements. The results were extended to assess the percentage of through vehicles in the right lane that would be impacted over a series of driveways. Single Driveways Traffic volume and impact characteristics at each study site were obtained. The right-lane volume ranged from about 245 to 820 vph, with an average of about 525 vph. The rightturn-in volume ranged from about 10 to 245 vph, with an average of about 100 vph. The percent of right-lane through vehicles impacted by right-turn-in ranged from about 2 to over 45 percent, with an average of about 17 percent. Figure 1 plots the percent of right lane through vehicles impacted as a function of right-turn-in volumes. A good linear relationship exists with a coefficient of determination (R 2 ) of 0.78. The percentage of right lane through vehicles impacted was about 0.18 times the right-turn-in volume. Four classes of right-turn-in volumes were identified with the following impacts at individual driveways. Right-Turn-In Volume (vph) Percent of Right Lane Through Vehicles That Were Impacted by Right Turn In Multiple Driveways <30 2% 31 60 7% 61 90 12% Over 90 22% The percentage of through traffic that would be impacted over a series of driveways in a quarter-mile road section was derived by extending the preceding analysis through probability analysis. The percentage of right-lane through vehicles being impacted at least once per quarter mile was derived as follows. The probability of being impacted is P 1. The probability of the complement, not being impacted, is 1 P 1. The probability of not being impacted for n driveways is (1 P 1 ) n. If n is the number of driveways per quarter mile, (1 P 1 ) n is the probability of not being impacted for a quarter-mile segment. The complement of this, the probability of being impacted at least once per quarter mile, P r, is then 1 (1 P 1 ) n. The results of these calculations are given in Figure 2. These values are independent of speeds since they deal only with the percent of right-lane through vehicles impacted

C-3 / 4 TRB Circular E-C019: Urban Street Symposium 50% P 1 = 0.182 R - 0.947 Coeff. of Determination = 0.78 45% % of Right Lane Thru Vehicles Impacted by Right-Turn-In, P 1 40% 35% 30% 25% 20% 15% 10% 5% 0% 0 25 50 75 100 125 150 175 200 225 250 275 FIGURE 1 Right-Turn-In Volume (vph) Single driveway case: Impacts vs. right-turn-in-volume. not how far back the impact area extends. Thus, if there was a driveway spacing of 100 feet (i.e., 13.2 driveways per quarter mile) and a right-turn volume of 30 to 60 vph, about 64 percent of the through vehicles would be impacted. If the driveway spacing was increased to 400 feet, 23 percent of the through vehicles would be impacted. Driveway Impact Lengths The information gathered from the 22 sites was analyzed to identify key patterns of driver behavior. Frequency and cumulative frequency distribution curves were prepared of impact lengths for each site. Figure 3 gives a composite picture of the cumulative distributions of impact length for all sites. (The x-axis gives impact length and the y-axis gives the percent of impacted vehicles that are impacted beyond a specified length.) However, more important than the distribution of impacts lengths for impacted vehicles is the distribution of impact lengths expressed in terms of the percentages of all

Gluck, Haas, Mahmood, and Levinson C-3 / 5 100% Percent of Right Lane Through Vehicles Impacted at Least Once per Quarter Mile 90% 80% 70% 60% 50% 40% 30% 20% 10% Right-Turn-In Vol per Driveway, R > 90 vph 60 < R < 90 30 < R < 60 R < 30 0% 100 200 300 400 500 600 700 800 900 1000 1100 1200 1300 Driveway Spacing FIGURE 2 Multiple driveway case: Vehicles impacted at least once per quarter mile. right-lane through vehicles, whether impacted or not. Four curves were derived for the four classes of right-turn-in volumes. The composite cumulative frequency distribution of impact lengths (Figure 3) was multiplied by the percent of right-lane through vehicles that were impacted by right-turn-in movements for each class. The results are shown in Figure 4 for the four ranges of right-turn-in volumes. The curves can be used to estimate the percentage of through vehicles in the right lane that would be impacted by right-turn-in traffic for various distances from a driveway for each range of right-turn-in volumes. Thus, beyond a distance of 150 feet upstream from the driveway entrance, for a right-turn-in volume greater than 90 vehicles per hour, about 7 percent of the right-lane through vehicles would be impacted. Beyond a distance of 100 feet upstream of a driveway, for a right-turn-in volume of 60 to 90 vehicles per hour, almost 7 percent of the right-lane through traffic would be impacted. Beyond a distance of 200 feet, about 2 percent would be impacted.

C-3 / 6 TRB Circular E-C019: Urban Street Symposium 100% Percent of Impacted Vehicles That Are Impacted Beyond the Specified Length 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 Impact Length (feet) Note: Arterial posted speeds ranged from 30 to 45 mph for all sites, with an average of 35.6 mph. The relationship between speed and impact length is discussed later. FIGURE 3 Cumulative frequency distribution of impact lengths for all impacted vehicles: Composite for all sites. Driveway-Influence Lengths The influence lengths or distances associated with various right-turn-in volumes and driveway spacings were also established. They consist of three components: impact length, car length, and perception-reaction distance: 1) The impact length was determined from field observations. 2) The car length was added because the field observations of impact lengths were taken at the front of each car, and the influence length should be measured to the rear of a vehicle. A value of 25 feet was used.

Gluck, Haas, Mahmood, and Levinson C-3 / 7 25% Percent Of Right Lane Through Vehicles That Are Impacted Beyond the Specified Distance 20% 15% 10% 5% Right-Turn-In Vol., R > 90 vph 60vph < R < 90vph 30vph < R < 60vph R < 30 vph 0% 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 FIGURE 4 Distance from Study Driveway (feet) Single driveway case: Cumulative frequency distribution of impact lengths (posted speed 30 mph). 3) The perception-reaction distance was based on a two-second reaction time that is typical of suburban conditions. (This represents the average of the 1.5 seconds and 2.5 seconds that AASHTO specifies for urban and rural conditions, respectively.) The equation for perception and reaction distance is: d = 147. St (1) where: d = perception/reaction distance in feet 1.47 = the conversion factor from miles per hour to feet per second

C-3 / 8 TRB Circular E-C019: Urban Street Symposium S = the speed in miles per hour t = the reaction time in seconds. A vehicle was considered to be influenced at or beyond another driveway if the influence length was greater than or equal to the driveway spacing minus the driveway width. Figure 5 shows the situation where the vehicle is not influenced at or beyond Upstream Driveway D'way Width Vehicle at Start of PIEV Car Length Driveway Spacing PIEV Influence Length Vehicle at Moment of Impact Impact Length Study Driveway When (Influence Length) < (Driveway Spacing) (Driveway Width), spillback does not occur. FIGURE 5 Influence length.

Gluck, Haas, Mahmood, and Levinson C-3 / 9 another driveway (i.e., influence length is less than the driveway spacing minus the driveway width). Influence distances were computed based on an average running speed of 30 mph. (Running speed is the travel distance divided by running time the duration during which a vehicle is in motion.) The resulting influence length (in feet) is: Influence Length = Impact Length + Car Length + PIEV distance (2a) Influence Length = Impact Length + 25 + 30( 2)( 1. 468) = Impact Length + 113 feet (2b) Figure 6 shows the cumulative frequency distribution of influence lengths for the four right-turn-in volume groups. It is similar to Figure 4, except that the curves are shifted 113 feet to the right to account for the above calculation. The posted speeds at the study sites ranged from 30 to 45 mph, with an average of 35.6 mph. Therefore, to be conservative, this figure was considered for posted speeds of 30 mph. Multiple Driveways The influence length curves were expanded to assess the effects of multiple driveways. The results are shown in Table 1 for a posted speed of 30 mph. The first two columns in this table show driveway spacing in increments of 25 feet, and the corresponding number of driveways per quarter mile (defined as n ). For each of the four right-turn-in volume per driveway categories, P 2 is the probability of a vehicle being influenced at or beyond another driveway for a single driveway condition. Figure 6 is used to get the P 2 values for any given driveway spacing. Thus, for a 225-foot influence length (30 mph) with more than 90 entering right turns per driveway, 10 percent of the through vehicles would be influenced beyond this distance (Figure 6). When 30 feet are deducted for the driveway width and the figure is reentered at an influence length of 195 feet, the corresponding value for a single driveway is 14.7 percent. This is the value that is entered as P 2 in Table 1 for R > 90 vph and a driveway spacing of 225 feet. The probability of a right-lane through vehicle being influenced by right-turn-in at least once per quarter mile is 1 (1 P 2 ) n. Effects of Speed The influence distance increases as speed increases. This is because driver behavior is keyed to separation in time (as well as space) and because perception-reaction distance increases as speed increases. The analysis found that the impact length was related to speed and the distance from the upstream traffic signal. (a) Impact Length Changes. The analysis established the following relationship between speed, distance from upstream traffic signal, and impact length. [ ] + + = L = 0. 361 ( s 30) 2 + s 0. 050d 86. 073 ( R 2 0. 778) (3)

C-3 / 10 TRB Circular E-C019: Urban Street Symposium 25% Percent Of Right Lane Through Vehicles That Are Influenced Beyond the Specified Length, P 2 20% 15% 10% 5% Right-Turn-In Vol., R > 90 vph 60vph < R < 90vph 30vph < R < 60vph R < 30 vph 0% 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 Influence Length (feet) Note: Influence length is equal to the impact length plus the PIEV distance plus the car length. FIGURE 6 Single driveway case: Cumulative frequency distribution of influence lengths (posted speed 30 mph). where: L = the mean impact length in feet. s = the running speed in mph (s 30 mph). d = the distance in feet from the nearest upstream traffic signal. This equation was used to convert the mean impact length for any percentile from a running speed of 30 mph to other speeds. The running and posted speeds were considered to be comparable for purposes of calculating impact lengths and influence areas. Solving the equation for d yields a value of 1,142 feet. Substituting different speed values into the equation while holding d constant yields their corresponding mean impact lengths. Dividing these numbers by 154 feet, the mean impact length for a posted speed of 30 mph gives a factor for converting impact lengths at any percentile

Gluck, Haas, Mahmood, and Levinson C-3 / 11 D'way Spacing (ft) TABLE 1 Spillback Rates: Percentage of Right Lane Through Vehicles Influenced at or Beyond Another Driveway (Posted Speed 30 mph) No. of D'ways per 1/4 Mi., n Single D'way, P 2 Right-Turn-In Volume per Driveway, R (vph) R < 30 30 < R < 60 60 < R < 90 R > 90 Multiple D'ways, At Least Once per 1/4 Mi., 1 - (1 - P 2 ) n Single D'way, P 2 Multiple D'ways, At Least Once per 1/4 Mi., 1 - (1 - P 2 ) n Single D'way, P 2 Multiple D'ways, At Least Once per 1/4 Mi., 1 - (1 - P 2 ) n Single D'way, P 2 Multiple D'ways, At Least Once per 1/4 Mi., 1 - (1 - P 2 ) n 100 13.2 2.4% 27.3% 7.5% 64.2% 12.2% 82.1% 21.8% 96.1% 125 10.6 2.4% 22.5% 7.5% 56.0% 12.2% 74.7% 21.8% 92.5% 150 8.8 2.4% 19.0% 7.5% 49.4% 12.2% 68.1% 21.7% 88.4% 175 7.5 2.3% 16.1% 7.2% 43.1% 11.8% 61.1% 21.0% 83.1% 200 6.6 2.1% 13.0% 6.6% 36.1% 10.7% 52.6% 19.1% 75.3% 225 5.9 1.6% 9.1% 5.0% 26.2% 8.2% 39.6% 14.7% 60.7% 250 5.3 1.2% 6.0% 3.6% 17.8% 5.9% 27.6% 10.6% 44.7% 275 4.8 0.9% 4.0% 2.7% 12.2% 4.4% 19.3% 7.8% 32.3% 300 4.4 0.6% 2.8% 2.0% 8.6% 3.3% 13.8% 5.9% 23.5% 325 4.1 0.5% 1.9% 1.5% 5.9% 2.4% 9.4% 4.3% 16.3% 350 3.8 0.3% 1.2% 1.0% 3.7% 1.6% 6.0% 2.9% 10.5% 375 3.5 0.2% 0.7% 0.6% 2.2% 1.0% 3.6% 1.9% 6.4% 400 3.3 0.1% 0.4% 0.4% 1.4% 0.7% 2.2% 1.2% 3.9% 425 3.1 0.1% 0.3% 0.3% 0.8% 0.4% 1.3% 0.7% 2.3% 450 2.9 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% for a posted speed of 30 mph to impact lengths at the same percentile for any other speed. The results are shown below. Posted Speed (mph) Mean Impact Length (feet) Factor Speed Ratio 30 154 1.00 1.00 35 164 1.07 1.17 40 194 1.26 1.33 45 241 1.56 1.50 50 306 1.98 1.67 55 389 2.52 1.83

C-3 / 12 TRB Circular E-C019: Urban Street Symposium Thus, to transform the 30 mph impact length curve to that for any other speed, the impact length for each percentile should be multiplied by the factor given above. Alternatively, impact lengths could be estimated based on the ratio of the observed speed to 30 mph; these indices are also shown. (b) PIEV Distance and Car Length. The car length remains constant, but the PIEV distances increase with speed. The resulting values are as follows: Posted Speed (mph) PIEV Distance 30 88 35 103 40 117 45 132 50 147 55 161 Accordingly, a series of tables were prepared that show the likelihood of spillback across driveways for various posted speeds, right-turn lane volumes and driveway spacings. These tables provide a means of assessing impacts where a driveway is added, closed, or consolidated. This involves comparing the percentages of vehicles extending across a driveway for speed and right-turn volumes for the before and after conditions. The information may be used to identify the cumulative impact of decisions concerning driveway location and unsignalized access spacing. Comparison of Results Table 2 gives unsignalized driveway spacings for various speeds, right-turn-in volumes, and spillback percentages. (Spillback is defined as the incidence of a right-lane through vehicle being influenced at or beyond the driveway upstream of the analysis driveway. It occurs when the influence length is greater than the driveway spacing minus the driveway width. The spillback rate represents the percentage of right-lane through vehicles that experience this occurrence.) The table also shows the New Jersey and Colorado access spacing standards for comparative purposes. The more liberal the standard (i.e., the greater the percentage of vehicles influenced at or beyond another driveway), the shorter the required driveway spacing. Similarly, the lower the assumed right-turn volumes (average per driveway), the shorter the allowable spacing. APPLICATION OF FINDINGS The preceding analyses can be used to establish guidelines for deceleration lanes and unsignalized access spacing. Deceleration Lanes Right-turn deceleration lanes are desirable to remove turning vehicles from through travel lanes, thereby reducing speed differentials and minimizing delays to through vehicles.

TABLE 2 Unsignalized Access Spacing Comparisons Possible Spacing Guideline Based Upon Posted Existing Spacing Maximum Allowable Spillback Rate Speed Standards (ft) 2 % Allowed 5 % Allowed Limit New Jersey Colorado Right Turn In Vol. per D'way, R (vph) Right Turn In Vol. per D'way, R (vph) (mph) DOT DOT R < 30 30<R<60 60<R<90 R > 90 R < 30 30<R<60 60<R<90 R > 90 30 125 200 320 380 405 430 260 335 360 385 35 150 250 345 405 435 460 280 355 385 415 40 185 325 380 460 490 520 305 400 430 465 45 230 400 430 530 565 610 340 450 495 540 50 275 475 490 620 665 725 380 520 575 630 55 330 550 550 725 780 855 420 590 665 740 Possible Spacing Guideline Based Upon Posted Existing Spacing Maximum Allowable Spillback Rate Speed Standards (ft) 10 % Allowed 15 % Allowed Limit New Jersey Colorado Right Turn In Vol. per D'way, R (vph) Right Turn In Vol. per D'way, R (vph) (mph) DOT DOT R < 30 30<R<60 60<R<90 R > 90 R < 30 30<R<60 60<R<90 R > 90 30 125 200 220 290 320 355 185 260 295 320 35 150 250 235 310 345 380 190 280 315 355 40 185 325 250 340 380 420 195 305 345 390 45 230 400 270 380 430 485 195 340 385 445 50 275 475 285 425 490 560 195 380 435 510 55 330 550 290 480 550 645 195 420 490 580 Possible Spacing Guideline Based Upon Posted Existing Spacing Maximum Allowable Spillback Rate Speed Standards (ft) 20 % Allowed 25 % Allowed Limit New Jersey Colorado Right Turn In Vol. per D'way, R (vph) Right Turn In Vol. per D'way, R (vph) (mph) DOT DOT R < 30 30<R<60 60<R<90 R > 90 R < 30 30<R<60 60<R<90 R > 90 30 125 200 140 245 270 310 110 230 255 295 35 150 250 140 260 290 330 110 245 275 315 40 185 325 140 285 320 370 110 265 300 345 45 230 400 140 315 355 410 110 290 335 385 50 275 475 140 345 400 470 110 315 370 435 55 330 550 140 380 445 530 110 330 415 495

C-3 / 14 TRB Circular E-C019: Urban Street Symposium The percentage of through vehicles in the right lane that must slow down or change lanes provides one possible basis for establishing the need for right-turn lanes. For arterial right-lane volumes of 250 to 800 vph, the percentage of through vehicles impacted was about 0.18 times the right-turn volume. This results in the following impacts that can provide a basis for providing right turn lanes. % Right-Lane Through Right-Turn-In Volume Vehicles Impacted vph (Approximate) 0% 0 2% 10 5% 30 10% 60 15% 85 20% 110 A criterion of 2 percent impacted suggests a minimum right-turn volume of 10 vph, and may be applicable in certain rural settings. A criterion of 10 percent impacted suggests a minimum of approximately 60 vph. A criterion of 20 percent suggests a right-turn volume of 110 vph. The latter criterion may be applicable in certain urban areas. The length of the deceleration lane is a function of the impact length and storage requirements. Access Separation Distances Both operational and safety considerations should govern unsignalized access spacing. A third consideration is the access classification of the roadways involved. Figure 7 compares the access separation distances for a range of spillback rates with the standards for Colorado and New Jersey, and AASHTO standards. Except for posted speeds of less than 40 mph, the resulting values for spillback rates of 10 to 20 percent fall between the New Jersey and Colorado (AASHTO safe stopping sight distance) criteria. Access separation distances, based on an average driveway volume of 30 to 60 vph, are shown in Table 3 for spillback rates of 5, 10, 15, and 20 percent. For the lower speeds of 30 and 35 mph, the access separation distance shown is based on the safety considerations major roadways should not have more than 20 to 30 connections per mile (both directions). For a posted speed of 40 mph, the access spacing would range from 285 feet to 400 feet depending upon which spillback rate was selected. For a posted speed of 50 mph, the access spacing would range from 520 feet to 345 feet depending on the spillback rate. Policy Guidelines Access separation distances should be established as part of access management programs, route retrofit plans, and community zoning ordinances. Ideally, direct property access from strategic and principal arterials should be provided only where reasonable access cannot be provided from other roadways. Whenever access is provided, spacing should be adequate to maintain safety and minimize impacts. The separation distances set forth in this paper can provide guidance in this effort.

700 AASHTO Calculated Stopping Distance* 600 Colorado DOT New Jersey DOT 500 5% Spillback Rate 30<R<60** 10% Spillback Rate 30<R<60** Driveway Spacing (feet) 400 300 15% Spillback Rate 30<R<60** 20% Spillback Rate 30<R<60** 200 * 9 fps 2 deceleration; 2.5 sec. perception-reaction time 100 ** Spillback rate is % of through vehicles influenced at or beyond another driveway at least once per quarter-mile. 0 15 20 25 30 35 40 45 50 55 60 65 Posted Speed (mph) FIGURE 7 Comparison of access separation criteria.

C-3 / 16 TRB Circular E-C019: Urban Street Symposium TABLE 3 Access Separation Distances (Feet) Based on Spillback Rate* Posted Speed(mph) SPILLBACK RATE** 5% 10% 15% 20% 30 335 265 (a) 210 (b) 175 (c) 35 355 265 (a) 210 (b) 175 (c) 40 400 340 305 285 45 450 380 340 315 50 520 425 380 345 55 590 480 420 380 (a) Based on 20 driveways per mile. (b) Based on 25 driveways per mile. (c) Based on 30 driveways per mile. *Based on an average of 30 to 60 right turns per driveway. **Spillback occurs when a right-lane through vehicle is influenced by right-turn-in to or beyond a driveway upstream of the analysis driveway. The spillback rate represents the percentage of right-lane through vehicles experiencing this occurrence. ACKNOWLEDGMENTS This research was performed under NCHRP Project 3-52 by Urbitran Associates in association with Herbert Levinson, S/K Transportation Consultants, and Philip Demosthenes. The authors of this paper want to thank Dr. Vergil Stover and Frank Koepke for their insights and suggestions. REFERENCES 1. Gluck, J., Levinson, H. S., and Stover, V. NCHRP Report 420: Impacts of Access Management Techniques, TRB, National Research Council, Washington, D.C., 1999. 2. Major, I. T., and Buckley. Entry to a Traffic Stream. Proceedings of the Australian Road Research Board, 1962, pp. 206 228. 3. Cribbins, P. D., Horn, J. W., Beeson, F. V., and Taylor, R. D. Median Openings on Divided Highways: Their Effects on Accident Rates and Level of Service. In Highway Research Record 188, Highway Research Board, National Research Council, Washington, D.C., 1963, pp. 140 157. 4. Treadway, T. B., and Oppenlander, J. C. Statistical Modeling of Travel Speeds and Delays on a High-Volume Highway. In Highway Research Record 199, Highway Research Board, National Research Council, Washington, D.C., 1967, pp. 1 18. 5. Berg, W. D., and Anderson, J. C. Analysis of the Tradeoff Between Level of Land Access and Quality of Service Along Urban Arterials. In Highway Research Record 453, Highway Research Board, National Research Council, Washington, D.C., 1973.

Gluck, Haas, Mahmood, and Levinson C-3 / 17 6. Bochner, B. Regulation of Driveway Traffic on Arterial Streets, Public Works, October, 1978. 7. Reilly, W. R., Harwood, D. W., Schuer, J. M., Kuehl, R. O., Bauer, K., and St. John, A. B. Capacity and Service Procedures for Multi-lane Rural and Suburban Highways, Final Report NCHRP 3-33, JHK & Associates and Midwest Research Institute, May 1989. 8. Special Report 209: Highway Capacity Manual (1994 update). TRB, National Research Council, Washington, D.C., 1994. 9. British Columbia Economic Analyses Guidebook. Planning Services Branch, Ministry of Transport and Highway, 1992. 10. Garber, N. J., and White, T. E. Guidelines for Commercial Driveway Spacing on Urban and Suburban Arterial Roads. In Conference Proceedings, Second National Conference on Access Management, Vail, Colorado, August, 1996. 11. McShane W. R. Access Management and the Relation to Highway Capacity and Level of Service, Technical Memorandum on Activity 4: Final Report, Task Work Order 14F, RS&H Project 992-1062-001, July 1996, prepared for Florida Department of Transportation. 12. McShane, W. R., Choi, D. S., Eichin, K., and Sokolow, G. Insights into Access Management Details Using TRAF-NETSIM, Conference Proceedings, Second National Conference on Access Management, Vail, Colorado, August, 1996.